Skip to main content

Graduate School

Retracting a PhD Degree, and Heading to Court

The circus is coming to town – that’s the only conclusion I can immediately draw from this post at Retraction Watch. It provides an update on something I blogged about last in 2014, the fallout from a retracted paper from the Martin group at UT-Austin. When last heard from, the university was trying to retract the PhD degree granted to Suvi Orr, a co-author on the paper. They did go ahead and do that, Orr sued, and UT reinstated the degree.

But now they’re taking another shot at retracting it, and Orr is back in court with them. The Austin American-Statesman has the story:

The lawsuit, filed Thursday, said UT officials revoked her degree but promptly reinstated it for “a do-over” during the first round of litigation. The suit contends that UT now plans to subject her to a “kangaroo court” whose members include undergraduate students lacking the expertise to interpret scientific data stemming from her research involving synthesis of chemicals. . .

Retraction Watch has a PDF available of the lawsuit as filed, and yikes. It accuses UT of basically throwing everyone else under the bus to protect Prof. Martin, and blames the third author (a post-doc) for what went wrong. This post-doc, the suit says, took over the project after Orr left with her degree:

A paper was published in which Prof. Martin was the leading author, and the post-doc and S.O. were coauthors of the journal article. S.O. intended and expected that the post-doc would fully characterize the compounds as part of reproducing her work. Unbeknownst to S.O., the post-doc instead made markings, scanned, and submitted experimental data containing S.O.’s file names and some incorrect (and inaccurate) pages as part of the journal article. On relevant issues related to Compounds P, Q, and R, he reached the same or similar conclusions as S.O. that are discussed in his post-doctoral final report. Nonetheless, the article was ultimately retracted, and the decision to retract was apparently made by Prof. Martin without input from his co-authors (S.O. and the post-doc).

The suit goes on to accuse Martin of somehow not encouraging Orr from obtaining an X-ray crystal structure of a key intermediate (this part is rather odd-sounding), and includes this allegation about the fallout from the published article:

Before the investigation was launched, Prof. Martin told S.O. in a phone conversation that there was a structural mistake in the journal article they published and that he would probably have to retract the article. He did not inform her that he had made a complaint against her. After S.O. learned that there was an investigation against her regarding the journal article she asked Prof. Martin about the details or meaning of the investigation. Prof. Martin told her to co-operate and admit misconduct to speed up the process.

UT is not commenting, as one would expect. The hearing is set to take place a month from now, and there will be quite a path to get to it, from all appearances. More details as they become available – if they do, because this looks, to my inexpert legal eyes, to have “out of court settlement” all over it. We’ll see.

40 comments on “Retracting a PhD Degree, and Heading to Court”

  1. Anonymous Researcher snaw says:

    My PhD was granted over 20 years ago, and nobody to my knowledge has ever questioned anything in my dissertation. However, to the best of my knowledge nobody beyond my committee, friends, and relatives has even read it. So I think my degree is safe from revocation.

    1. Dana says:

      Yes, we used to joke about putting a 10 dollar bill in the library copy of our PhD thesis (maths) in the library – would still be safely found there when visiting again in 20 years… Not a lucrative investment, but a safe one.

    2. David says:

      Same here, minus the family and friends.

  2. exGlaxoid says:

    I’m not even sure that my committee members read mine… And I am sure that my friends did not. What a waste of paper.

  3. blame the PI says:

    and in that story, as usual the PI is clean. Can’t excuse someone from cheating (the student), but the PI’s name is on the paper (supposing he read it, also as a PI questioned the student during group meetings / one on ones), and signed the dissertation (supposing he’s read the dissertation)…

  4. Hap says:

    I think people are willing to accord a lot less blame to a PI if the coworkers in question falsified the work (at least the first time – I assume a PI would be not looking (much) for evidence that their students are lying, but more for evidence that something strange in the chemistry is going on) and if they did not assist or cover it up (e.g., Sames’ firing the grad students unable to reproduce Sezen’s work). Most of the credit the PI would have received would have gone away when the paper was retracted, so he’s not getting a whole lot of credit without blame.

    I don’t know what’s going on, though – the grad student’s story sounds like two bags full, but if UTexas had evidence of falsification, then it’s well past time to show it – not having done so (with ample opportunity) makes it look like it has something to hide, I also didn’t realize that show trials were back in fashion (although maybe the funny student/faculty assignment is a preexisting requirement?)

    1. Phil says:

      If she didn’t falsify results (and looking at the SI and her thesis,* there appears to be no evidence that this is the case), what grounds does UT have to revoke her degree? I am asking myself the same question, what evidence of falsification could exist, and why haven’t they produced it yet?

      If a Ph. D. can be revoked for structural misassignment, there are quite a few natural products isolation chemists who should be worried. If a Ph. D. can be revoked because others fail to reproduce one or two of your experiments, pretty much everyone in synthetic chemistry should be worried (not to mention all those biologists to whom Amgen is throwing down the gauntlet).

      * Pages 177-195 of her thesis contain plenty of blood, sweat and tears. And in the end, the molecule wasn’t completed. It just doesn’t seem like making up those results would be worth it.

      1. Hap says:

        Falsification would be the only reason I could see why they could revoke a degree (it was awarded under pretenses the awardee knew were false), unless it were really gross incompetence, in which case more than a little blame ought to fall on the PI. Not reproducing experiments seems insufficient (why we don’t publish failed repetitions very much – how do I know you couldn’t reproduce my work because you weren’t good enough and not because I lied or because I insufficiently specified conditions?) , and yes, that seems like a standard lots of good chemists could fail easily. I don’t know how bad gross incompetence would be to revoke a degree.

        At this point, either they’d better produce evidence of fraud or cash for a libel suit payout. It’s hard to come up with a good reason why UT hasn’t disclosed their evidence (other than it can’t stand the light of day), and if that’s the reason, and they’re pushing a show trial anyway, there’s something else going on that needs to be public.

        1. Phil says:

          Lots of good chemists, including the best.

  5. dearieme says:

    I was once external examiner of a PhD dissertation that turned out to be an inferior copy of part of a dissertation written by one of my own students some years before. The last I heard of it was that the supervisor of the student concerned had been given sick leave, and that the student was going to resubmit under the supervision of someone else.

    The fact that I was appointed external examiner suggests to me that his home university already feared for the competence or sanity of the supervisor concerned.

  6. Chrispy says:

    Well if a settlement comes out of it then she’ll have gotten more from her PhD than a lot of others…

  7. hook em' says:

    I don’t think the circus ever left the University of Texas at Austin. It certainly was in full swing during the 90s, at least in the Department of Chemistry. That era featured major turf wars between the chemistry faculty, including faculty deployment of contact explosives (nitrogen triiodide) to retaliate and deter faculty-sponsored government audits (of federal grants), complete with routine visits from the Austin fire department. There was also the case of the UT stockroom supplying a meth lab on the outskirts of town with pharmaceutical grade reagents and equipment (chemistry hoods included). Good times, way ahead of the curve of Breaking Bad. In comparison, this PhD controversy seems pretty mundane.

    1. Dennis says:

      Huh. One of my professors during my undergrad mentioned there being several nitrogen triiodide incidents that he either observed or participated in during his graduate work at UT. I had always assumed he was making it up because it sounded too friggen crazy, but maybe I was wrong?

      1. hook em' says:

        Your prof was telling the truth – it really was a circus. There was significant “class participation” by some graduate students, either implicitly or explicitly condoned by faculty, that contributed to the various turf battles. Some ended up doing jail time because of their more destructive exploits. Of course, no faculty were ever penalized for being adolescent, but it probably made Alan J. Bard and Norman Hackerman shake their heads in dismay. At least it wasn’t a dull department and apparently it’s still crazy after all these years… Keep it Weird Austin!

  8. PorkPieHat says:

    Really, was this going to end well for the University? Or, is it going to end anywhere else but an “out of court settlement” that Derek speaks of? Gotta wonder why the University would pursue a retraction given those odds (they better have the goods on fraudulent actions by SO).
    Hey, “hook’em”, please tell more…this sounds really juicy, the stuff about UTA in the ’90s. There’s gotta be more where that came from.

  9. Me says:

    One of Orr’s allegations is that she was incorrectly supervised because she wasn’t taught how to recrystallize a molecule? Hmmmm…. apparently an X-ray structure would have solved it and they didn’t get one because the PI didn’t teach her how to crystallise something…..what a selfish PI. You’d have thought that after years as a med chemist she might have learned that there is no good way…… unless the good people at Pfizer have a method I haven’t heard of.

    For me, the numbers don’t add up in the paper. The question is down whose fault that is, and whether S.O. is to blame….

    1. Hap says:

      I don’t know if it matters if she was to blame – the lack of reproducibility of the results doesn’t seem like a good reason to revoke a degree unless there was fraud (and SO and not someone else was provably at fault), and in that case, UT’s going to have to show that in public. If SO’s PhD was based on one paper, well, her committee knew that six years ago, and still graduated her. If the failure of one paper (without evidence of misconduct) is sufficient to revoke the degree, then the school and the advisor messed up graduating her in the first place.

    2. chemist says:

      Honestly this section sounded like lawyer-ese to me. It doesn’t make much sense to anyone trained in chemistry that someone should have to crystalize every small molecule synthesized, and if 1D NMR is not sufficient to characterize the structure and the compound doesn’t readily crystalize, there is always 2D NMR. However, if there was a more cogent argument to be made to a group of non-experts, that might have been what the attorney came up with.

      One of the shadiest part of how they are dealing with this case is that the hearing will involve a panel of faculty in unrelated disciplines and undergraduate students. How is it at all appropriate for undergraduate students to determine whether or not she should have her PhD revoked?

    3. cop says:

      Yeah that part was a bit of a cop-out. I don’t know anyone who has been personally tutored in how to crystallize a compound by their PI. As a researcher, the best course of action is to go look up methods and try them out. If she is unable to achieve anything without being personally walked through by her PI (and let’s face it, that is all just lawyer talk) she probably shouldn’t have a PhD.

  10. Anonymous says:

    Rough stuff. The worst thing I ever saw was a committee yanking someone’s dissertation defense the Friday before the examination; everything was on track until the PI went on a sudden medical leave, and the previously supportive committee told them that the work was unacceptable. Not sure if leaving with a master’s then was better than having to fight to keep your PhD in court a few years later.

  11. Diabolo Menthe says:

    I know the external reviewer read my thesis because his first comment at the defense was that, given the numbers of typos and grammatical mistakes, it became obvious to him that none of my thesis’ supervisors bothered reading the thesis prior to submission (it was before automatic proofreading was included in writing software !)

  12. dave w says:

    Isn’t the fact that courts are even willing to hear such cases sort of a nasty precedent? Whatever the facts of this case (about which I remain confused), do we want to live in a world in which the granting (or otherwise) of a degree is subject to judicial review? (Should students be able to, e.g., sue their teachers because of a disappointing grade or evaluation – or should this be considered as a matter internal to the academy, with which the law does not intersect?)

    1. Hap says:

      Retracting a degree already given (and paid for, by someone) is somewhat different than influencing grades or degrees not yet given. (The costs of doing so probably prevent it from happening much.) It may also be different for Texas being a public school, in which case there may be procedural requirements in place (and responsibilities) not present at a public school.

      I imagine that the case for higher education would get significantly worse if schools could arbitrarily retract your degree with little oversight. In that case, given the choice between relying on an academic bureaucracy with not much accountability or the legal system, I probably would trust the courts more.

  13. Pedro Arrechea says:

    UT Austin has sent a rebuttal to the lawsuit…
    They accuse her of manipulating the NMR spectra for compounds 3.210, 3.237, and 3.238 in her dissertation.

    1. Phil says:

      That would be 26, 26, and 27 in the OL paper. The last six pages of the supplemental have 1H and 13C NMR spectra for those compounds. They aren’t particularly clean, but do you see evidence of manipulation? How does UT or Martin know the spectra were doctored?

      Maybe they are waiting until it goes to trial to present the evidence, but it seems like it should have been presented already.

      1. bad wolf says:

        Yeah, the spectra (particularly one 13C) aren’t clear but these are late-stage transformations on a 1-5mg scale, so…. i can imagine a finicky step that was low yield, maybe worked once or twice, and they just push ahead to finish up. To me, you’d better have better evidence than that, or there’s a whole mess of PhD’s out there living on borrowed time.

        I just think it’s such a nightmare to have someone call you up and tell you they’ve rescinded your degree, not sure anyone should be put through that. Besides, after working at Pfizer for 8 years, I think her supervisor there would be the best judge of whether she ‘deserves’ a PhD or not.

        1. Hap says:

          Assuming they found evidence of manipulation, how do you prove (or know) it was Orr and not the post-doc who she says was responsible for the spectra? Are the spectra data or otherwise marked so that you’d know when they were taken and by whom?

          This seems like a weak reed to revoke a degree on.

          1. Phil says:

            I forgot about that part. Orr claims the post-doc was responsible for the spectral data in the OL paper. Her thesis does not include any full spectra as far as I can tell, only peak reports. Those would be a bit easier to make up, but still, where is the evidence? Were these structures actually prepared afterward and the peaks don’t match? If so, why was the paper retracted?

          2. bad wolf says:

            Unless they have security camera footage of someone, i can’t imagine the “evidence” isn’t as weak as the accusations. As Phil asks, why was this paper retracted? I’m sure we all had colleagues we wish hadn’t been given degrees from our schools, but this whole thing doesn’t make much sense. Is there a patent hinging on the success of this synthetic sequence? How did this retraction preserve Martin’s reputation? Doesn’t this make Martin and UT look worse?

            If i could i would definitely spread this news around the prospective grad students for next year. “Come to the school that may decide to claw back your degree years later! Work for the PI who did that!”

  14. Tolen says:

    It’s funny how being a Professor is akin to an aristocratic rank. So when things go right, the Prof gets the glory and maybe even money?

    Yet if the data is shabby or even the student gets hurt, its not his or her fault? My experience is that most Profs work 40 hr weeks and take the weekends off.

    And a Post-doc is a student position. Ask any employer or even the govt with regard to your student loans! The Professor is 100% responsible.

    Yet another reason to eliminate tenure.

  15. Hap says:

    I imagine you could do audit trails, but in general schools don’t have e-notebooks (expensive for a group, or requiring time and competence and I don’t schools want to spend the overhead on infrastructure). I think some of the Sames-Sezen proceedings were based on NMR accounts (didn’t Sezen have, like, twelve of them?), but not most. To get the tabulated (or to get far enough that you would be tabulating them for anything useful) spectral data, I assume you’d have had the real spectra, and probably have showed it to your advisor or coworker.

    If it’s skillful enough to get by initial queries (no big blocks over baseline, ridiculously smooth baseline, peak copying, etc.), you’d figure that spectral manipulation might require some skill and development. If that’s the case here, why just these three spectra?

    The ratio of hat to cattle here seems low.

    1. Hap says:

      ratio of hat to cattle seems high. Ack,

  16. Crumpy says:

    Another graduate student from the Martin Group (Alexander Nichols – you can google the dissertation and look at the .pdf) seems to shed some light on this matter in his dissertation, where he tried to expand on Dr. Orr’s studies. Particularly p.177-178.

    1. Hap says:

      Thank you.

      1. Phil says:

        OK, I am convinced that Orr did not correctly assign the structure of the alleged RCM product. I also learned something new about ruthenium carbene chemistry, the proposed mechanism by which the pyrrole product is formed is pretty darned interesting (and, according to Nichols, unprecedented).

        But where is the falsification? Doesn’t Martin share some responsibility for the misassignment? If he didn’t believe the assignment, why did he sign off on the thesis in the first place?

        Also, what proved the misassignment was not an X-ray crystal structure, but 2D NMR experiments. So why are we talking about crystallization?

        1. Hap says:

          If it’s just a straight misassignment, why is UT pulling the degree? That doesn’t seem like an appropriate penalty to me.

          1. Crumpy says:

            That is still unclear, but at least we now know why the paper was retracted. Although, oddly the retraction notice states that the RCM was not reproducible, suggesting at one point it did work. This is instead of just saying it was an incorrect structural assignment. Who knows?

        2. Hap says:

          1) I assume Nichols wouldn’t go out of his way to frag his coworker, but the thesis doesn’t seem to imply anything funny, but simply that the product was misassigned. If there had been shenanigans, wouldn’t the language have been different?

          2) If the spectra for the byproduct had matched that of the desired product, I assume that they would have found its structure differently (after the next synthetic step, maybe, when they didn’t get what they thought, instead of stopping immediately and saying “We don’t have the right stuff.”). Given that, if Orr had manipulated its spectrum, you’d expect that the manipulated spectrum would have differed from that of the byproduct. The spectrum in her thesis, though, matched the byproduct. It doesn’t make sense that someone would have manipulated the spectrum to match the wrong product (and one for which there was not much precedent).

          Am I missing something?

          1. Phil says:

            Presumably, the vinyl group was hydrogenated and the ester was indeed saponified to the carboxylic acid. So, the NMRs are probably somewhat internally consistent (albeit messy as bad wolf points out). So those spectra are probably not falsified either, just misassigned based on the assumption that the RCM closed the ring rather than give the pyrrole (still not sure I understand how that works!).

  17. Marya Lieberman says:

    I served on a thesis committee where the student had attempted a multip step synthesis, which had gone awry in an early step where the intended product was a di-alcohol. Instead, an internal cyclization occurred. The product of the reaction was misidentified as the desired disubstituted product, taken through the whole synthetic scheme, and every inconsistency in the IR and NMR spectra was ignored or shoehorned into evidence for the hoped-for structures. There was no deliberate falsification involved, and no question of yanking the thesis. But it was a good demonstration of the human capacity for self-deception.

Comments are closed.